Programming 12f629

[stay_aliveuk]

Can any one help. I am trying to program a 12f629 with my new JDM programmer. It works with my collection of 16f84's and a new batch of 16f628's.

The error message when programming (using winpic800) is that it is not reading what it has just tried to program to the first address.

I initially thought it was my program but i have also tried a hex file from 'tinternet and still I am getting the same message.

I have also noticed that there is value is the last address position even on a brand new chip.

Finally, what is the bandgap, is the setting of this during programming critical?

Best regards

[ehoskins]

I'm not familiar with your programmer so will let someone else chime in who can help there, but regarding seeing data in the last address of the memory, that is the internal oscillator calibration byte put in by Microchip into all 12F629's. Check the data sheet. Do not overwrite that byte; or at least if you are going to erase the entire chip then take a note of that byte so you can write the value back to the chip later. If you're going to use the internal oscillator and the program uses critical timing you will want to invoke the oscillator calibration statement as explained in the PICBasic manual.

[Archangel]

Hi,
I also use a JDM programmer and it came with winpic800 and I was unable to get it to work on anything. I tried ICPROG (free from internet) and have used ever since. I never tried to program 12f series though. Oh and ICPROG looks almost exactly like the software you are using too.
Joe

[ardhuru]

This is a typical problem with JDM type programmers; if you have disabled MCLR *and* are using internal oscillator, you would have issues.

Here's what was found on another thread:

Quote-

Some of the programmers that are available apply VDD all the time. The 675 requires VPP to be applied before VDD in order to enter programming mode. Problems manifest themselves when you use the internal clock and disable MCLR. In this state, applying power will cause the programmed code to run and prevent programming mode being entered. A simple solution to this problem is to disconnect VDD and connect a 1k resistor from VPP (MCLR) to VDD and a 5.1V zener from VDD (cathode) to ground, thus ensuring VPP is applied first. Be aware that this mod may stop you programming some other pics like the 16F88.

Mike.
Nigel, A Friend of mine recently had problems with your tutorial. The problem turned out to be as described above. He programmed tutorial 1.2 and although WinPic said it had failed to program it, the data in line was pulsing and the code was running while the chip was still in the programmer. This also happened with IC-PROG. The programmer was the one published by Silicon Chip magazine. The above mod fixed it.

-Unquote

A better solution, of course, would be to build a parallel port based programmer.

Regards,

Anand

[mister_e]

or invest in a real device programmer... ICD-2, Elnec, PicStart, EPIC...etc etc

You have to be rich to buy cheap

[ardhuru]

or invest in a real device programmer... ICD-2, Elnec, PicStart, EPIC...etc etc

You have to be rich to buy cheap

Sure, but there are EPIC clone designs available in plenty to build one yourself. In fact, I have built one that is almost the same as EPIC, but even more reliable on account of certain minor differences (such as pull-ups on both sides of the 74 buffer, clock, data lines pulled up to *unswitched* 5 volts, and so on). Coupled with the Winpic800, its a robust team.

Regards,

Anand

[ehoskins]

Sure, but there are EPIC clone designs available in plenty to build one yourself. In fact, I have built one that is almost the same as EPIC, but even more reliable on account of certain minor differences (such as pull-ups on both sides of the 74 buffer, clock, data lines pulled up to *unswitched* 5 volts, and so on). Coupled with the Winpic800, its a robust team.

Regards,

Anand

and where are these EPIC clones? A link would be nice!

[ardhuru]

http://www.coolcircuit.com/project/picprog/

Reagrds,

Anand